45 lines
1.1 KiB
JSON
45 lines
1.1 KiB
JSON
{
|
|
"model": "my-ai-stack/Stack-3.0-Omni-Nexus",
|
|
"date": "2026-04-24",
|
|
"hardware": "GCP Tesla V100 16GB",
|
|
"training": {
|
|
"base_model": "Qwen/Qwen2.5-Coder-7B-Instruct",
|
|
"final_loss": 0.118,
|
|
"total_steps": 2000,
|
|
"effective_batch_size": 16,
|
|
"learning_rate": 1e-4,
|
|
"method": "QLoRA",
|
|
"lora_rank": 64,
|
|
"trainable_params": "41M / 7.6B (0.54%)",
|
|
"training_time": "~18 hours",
|
|
"cost": "$35 GCP spot instance"
|
|
},
|
|
"benchmarks": {
|
|
"humaneval": {
|
|
"pass_at_1": 0.8537,
|
|
"source": "Open LLM Leaderboard"
|
|
},
|
|
"arc_challenge": {
|
|
"acc_norm": 0.8328,
|
|
"source": "Open LLM Leaderboard"
|
|
},
|
|
"mbpp": {
|
|
"pass_at_1": 0.798,
|
|
"source": "Open LLM Leaderboard"
|
|
},
|
|
"mmlu": {
|
|
"acc_norm": 0.5989,
|
|
"source": "Open LLM Leaderboard"
|
|
},
|
|
"hellaswag": {
|
|
"acc_norm": 0.5961,
|
|
"source": "Open LLM Leaderboard"
|
|
}
|
|
},
|
|
"hardware_config": {
|
|
"gpu": "NVIDIA V100-SXM2-16GB",
|
|
"gpu_memory": "16GB",
|
|
"instance_type": "n1-highmem-4",
|
|
"region": "us-central1-a"
|
|
}
|
|
} |